A Novel Support Vector and K-Means based Hybrid Clustering Algorithm

摘要：

Data clustering is a hot problem and has been studied extensively. In this paper, we propose a novel support vector and K-Means based hybrid algorithm for data clustering. Firstly, we identify the outliers and overlapping data points through the support vector approach. Secondly, we remove the outliers and overlapping data points and then run the K-Means on the rest data points to obtain clustered data set. Finally, we build support vector description for each cluster, and then assign the removed data points to the cluster with the smallest distance, thus resulting in labeling the whole data set. Simulation results demonstrate that the proposed algorithm is effective, which exploits the advantages of both support vector clustering and K-Means.

关键词： Data Clustering Support Vector Clustering K-Means clustering

作者: Liang Sun Shinichi Yoshida Yanchun Liang

作者单位: College of Computer Science and Technology,Jilin University,Changchun,130012,China School of Informa School of Information,Kochi University of Technology,Kochi 782-8502,Japan College of Computer Science and Technology University,Changchun 130012,China

会议类型: 国际会议

会议名称: 2010 IEEE信息与自动化国际会议(ICIA 2010)

会议地点: 哈尔滨

会议语种:英文

页码: 1-5

在线出版日期: 2010-06-20（万方平台首次上网日期，不代表论文的发表时间）